While DeepSeek LLMs have demonstrated remarkable abilities, they don't seem to be devoid of their limits. Here are a few likely negatives of this kind of versions: None of the GPT-4o or Claude 3.5 Sonnets could response this straightforward issue properly. Only o1 was capable of finding the proper remedy https://x.com/kidtsang/status/1884008035535782292